Art & Design

Recently updated agent-skill-related GitHub repository: devndesigner6/tinyplace.

Codex skill for local image generation

A codex skill for transforming images into Van Gogh style oil painting

让纯文本模型更好地做视觉任务的DeepSeek Harness插件：带意图的图片问答、长截图 OCR、UI 还原等｜DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts, and Web UI.

Orchestrate disciplined AI agent swarms in tmux worktrees to forge reliable software engineering workflows.

Dudley Utility Services — brand & design system: dus-brand Claude skill, locked templates, and brand book

The open marketplace for agent skills

An open-source AI coding Skill that reads your repository before creating an editable 1280×640 GitHub Social Preview.

Agent skill for creating presentation decks with the Deckmaker CLI

Catalog & docs site for IncredAgents — production AI agent skills (SKILL.md) for Claude Code and compatible agents. Every skill gets its own page with real output demos.

Your durable workspace across AI agents.

Evidence-grounded paper analysis Skill for Codex and Claude Code, with offline HTML reports and Agent-learning datasets.

A research collaborator skill for Claude Code that flags its confidence, resists hallucination, and cites sources

Local, on-demand maintenance skills for SvelteKit + TypeScript repos — security, performance, refactoring, accessibility, and test coverage

Claude Code skills for authoring AGENTS.md, CONTEXT.md, and DESIGN.md

A seven-question post-ship retrospective skill for Claude Code, built around confidence, blind spots, and decay

Drafts and reviews prose for Claude Code — writes new content in the right voice per channel, or scans existing text for AI patterns and rewrites it in an authentic human voice

Installable agent skill for safely configuring FoloToy character personas, opening lines, and voices

Claude skill: stylize photo into lo-fi doodle style

Evidence-led, Vanguard-policy-safe engineering guidance Skill for League of Legends data integrations. Build compliant tools better.

y1n-flow · 多模型编排开发流 | A multi-model orchestrated dev workflow for DeepSeek Harness:主代理编排、子代理执行,plan document 为唯一事实源,内置编码验证闭环 / lead agent orchestrates, sub-agents execute, one plan doc as source of truth, built-in review loop

中文内容创作 SKILL

GLEIF vLEI 信任鏈的可執行沙盒 — A runnable model of the GLEIF verifiable LEI trust chain: real Ed25519, CESR, SAIDs and ACDC edge operators in pure Python, plus a path to a real KERIA stack.

Claude Code skill: build an interactive map of how a topic works in an unfamiliar codebase